Yau Jim Yip

Automatic Segmentation for Breast Skin-line

The accurate segmentation of breast in mammograms is critical for computer-aided diagnosis (CAD) of breast cancer. The challenges of breast segmentation come from tissue superimposition, low contrast and low visible breast boundary, noise areas and artifacts. In this research, a novel Automatic Breast Segmentation (ABS) approach was proposed. The experiment results showed that the proposed ABS approach can segment breast boundary reliably and accurately.

Mobile agent based data fusion for wireless sensor networks with a XML framework

A device for preventing a motor from being locked comprises a position indicator for indicating a position to be moved in response to rotation of the motor, a position detector for detecting the position being moved in response to rotation of the motor, a servo control unit for supplying a current to the motor so that an output signal from the position detector will be equalized with an indication by the position indicator, a stop detector for detecting stoppage of the motor in response to a differential of the output signal from the position detector, and a cutoff control unit for cutting off the current supplied from the servo control unit to the motor in response to an output signal from the stop detector.

Automatic Detection of Pectoral Muscle with the Maximum Intensity Change Algorithm

The accurate segmentation of pectoral muscle in mammograms is necessary to detect breast abnormalities in computer-aided diagnosis (CAD) of breast cancer. Based on morphological characteristics of pectoral muscle, a corner detector and the Maximum Intensity Change (MIC) algorithm were proposed in this research to detect the edge of pectoral muscle. The initial result shows that the proposed approach detected pectoral muscle with high quality.